The Research And Application Of CAVLC Improvement Algorithm Based On H.264

With the continuous development of multimedia technology, H.264video codingstandard as a new generation appeared in people’s eyes. Compared with the previous videocoding standard, H.264has better performance on video data compression efficiency andnetwork adaptability. However, the algorithm complexity of H.264eneoding and decoding ishigher, to some extent, which affects the application of H.264, especially in some real-timeenvironment. Context-based Adaptive Variable Length Coding (CAVLC) is advanced entropycoding technique used in H.264to improve the data compression efficiency due to their use ofcontext information, but it also reduces the encoding and decoding speed of H.264.Optimization on the algorithm of CAVLC entroy decoding method can speed up the H.264eneoding/decoding and promote its application. The main content of this paper is divided intotwo parts.Part I: First, the paper describes the basic principle and decoding process of H.264videocoding standard, and the principle of entropy decoding. After that, the paper presents theoptimization of the number of nonzero coefficients (TotalCoeffs) and the number of trailingcoefficients (TrailingOnes). Then, by doing statistical analysis and optimizing the grouping forthe code table, it can reduce code stream match time, improve the decoding speed. Finally, thepaper implements improved CAVLC and compares for JM and FFmpeg, which are the popularplatform of open source decoders of H.264.Part II: Because of the wide application of the Android operating system, it’s veryimportant to develop a good performance of the video player on mobile devices. Theunmodified CAVLC decoding on FFmpeg expended a lot of memory space, so that it is notconducive to the phone’s real-time decoding. By transplanting FFMpeg codec library to theAndroid platform, the paper designed and implemented a video player which supportedmultiple video formats with improved CAVLC decoding. There are some basic functionsincluding menu display, playback, pause, stop and auto-hide and so on. By the testing, thesystem validated the feasibility of improved algorithm on the mobile Android platform.